• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of eukcellPermissionsJournals.ASM.orgJournalEC ArticleJournal InfoAuthorsReviewers
Eukaryot Cell. Jan 2006; 5(1): 26–44.
PMCID: PMC1360252

Genome-Based Approaches to Understanding Phosphorus Deprivation Responses and PSR1 Control in Chlamydomonas reinhardtii


The Chlamydomonas reinhardtii transcription factor PSR1 is required for the control of activities involved in scavenging phosphate from the environment during periods of phosphorus limitation. Increased scavenging activity reflects the development of high-affinity phosphate transport and the expression of extracellular phosphatases that can cleave phosphate from organic compounds in the environment. A comparison of gene expression patterns using microarray analyses and quantitative PCRs with wild-type and psr1 mutant cells deprived of phosphorus has revealed that PSR1 also controls genes encoding proteins with potential “electron valve” functions—these proteins can serve as alternative electron acceptors that help prevent photodamage caused by overexcitation of the photosynthetic electron transport system. In accordance with this finding, phosphorus-starved psr1 mutants die when subjected to elevated light intensities; at these intensities, the wild-type cells still exhibit rapid growth. Acclimation to phosphorus deprivation also involves a reduction in the levels of transcripts encoding proteins involved in photosynthesis and both cytoplasmic and chloroplast translation as well as an increase in the levels of transcripts encoding stress-associated chaperones and proteases. Surprisingly, phosphorus-deficient psr1 cells (but not wild-type cells) also display expression patterns associated with specific responses to sulfur deprivation, suggesting a hitherto unsuspected link between the signal transduction pathways involved in controlling phosphorus and sulfur starvation responses. Together, these results demonstrate that PSR1 is critical for the survival of cells under conditions of suboptimal phosphorus availability and that it plays a key role in controlling both scavenging responses and the ability of the cell to manage excess absorbed excitation energy.

While phosphorus (P) is an abundant element in the Earth's crust, its availability can limit the growth of organisms present in both aquatic and terrestrial environments. P is essential for many fundamental processes that sustain life, including nucleic acid synthesis, membrane synthesis, energy metabolism, signaling, redox reactions, and modification of protein activities. The major form of P readily assimilated and utilized by most organisms is the phosphate anion (Pi). While available Pi (soluble Pi) is generally present at concentrations of <10 μM, most organisms require cellular Pi concentrations in the millimolar range (for reviews, see references 4, 40, and 45). For this reason, Pi supplementation is necessary to boost crop production and is included as a major constituent of fertilizers. However, following the introduction of Pi into soils, it can be rapidly precipitated as insoluble salts, become adsorbed to soil particles, or leach from agricultural fields, contaminating both groundwater and aquatic ecosystems. Manipulation of crop species to improve their ability to mobilize soil Pi may result in increased agricultural yields and a slowing of the consumption of exhaustible Pi sources.

Eukaryotic signal transduction pathways for sensing and responding to P starvation have been best studied in the yeast Saccharomyces cerevisiae. P-deficient S. cerevisiae cells increase their capacity for Pi uptake by expressing a high-affinity Pi transport system, mainly the Pho84p H+/Pi symporter, which operates at acidic pHs (3, 6). The Pho89p protein is a Na+/Pi cotransporter and may be utilized at more alkaline pHs (29). Pi that is covalently bound to organic molecules may be mobilized by cleavage by an acid phosphatase encoded by the PHO5 gene (reviewed by Vogel and Hinnen [50] and by Oshima [36]). There are many genes that are coregulated under P starvation conditions in S. cerevisiae, forming the PHO regulon (35). Changes in gene expression that accompany P starvation of S. cerevisiae are regulated by a system comprised of a cyclin (Pho80) and a cyclin-dependent kinase (Pho85). Under P-replete conditions, the kinase activity of Pho80/Pho85 catalyzes hyperphosphorylation of the Pho4 transcription factor, which is then excluded from the nucleus. When P becomes limiting, the cyclin-dependent kinase inhibitor Pho81 is activated, preventing Pho80/Pho85 kinase activity and leading to hypophosphorylation of Pho4 and its subsequent import into the nucleus. Once in the nucleus, Pho4 and Pho2 interact, populating the promoters of target genes and activating gene expression (reviewed by Lenburg and O'Shea [26]).

In photosynthetic eukaryotes, Pi is also critical for chloroplast function. Pi must be transported from the cytoplasm of the cell, across the double membrane of the plastid envelope, and into the chloroplast stroma, where it is used to satisfy the biosynthetic and energetic requirements of the organelle. Pi is needed for both DNA and RNA synthesis within the chloroplast, as the plastids contain their own genomes and ribosomes, and each chloroplast usually has multiple chromosome copies. Photosynthetic generation of ATP and the synthesis of phospholipids represent other critical chloroplast activities requiring Pi, and the phosphorylation of polypeptides of the photosynthetic apparatus modulates the properties of light-harvesting functions in response to changing environmental conditions (55). Furthermore, P limitation leads to depletion of the pool of phosphorylated intermediates in the pentose-phosphate cycle, which results in a marked reduction in photosynthetic carbon fixation (5, 22). Studies with Chlamydomonas reinhardtii have demonstrated that P and sulfur deprivation causes a similar loss of photosynthetic electron transport activity, the consequence of a combination of reduced photosystem II (PS II) abundance, accumulation of PS II QB nonreducing centers, an increase in nonphotochemical quenching, and an increase in the tendency of the cells to be in state II (53). These processes all redirect absorbed excitation energy away from PS II and linear electron transport. The observations described above support the idea that when nutrient availability limits the growth of the cell, a suite of mechanisms are mobilized to manage/dissipate excess absorbed light energy, reducing the potential of the system to generate reactive oxygen species which could cause severe cellular damage.

Little is known about the regulatory pathways by which plants and algae sense and respond to P deficiency. Genetic screens for C. reinhardtii mutants with a diminished Pi uptake capacity or reduced alkaline phosphatase activity during P-limited growth (46, 54) led to the discovery of the phosphorus starvation response 1 (PSR1) gene, which encodes a nucleus-localized factor with a MYB-like DNA binding domain and a coiled-coil protein-protein interaction domain (54). PSR1 is required for normal growth and the synthesis of alkaline phosphatase following starvation of the cells for P. Furthermore, the abundance of the PSR1 transcript and its protein product also increases during P starvation (54). Subsequent studies with Arabidopsis thaliana revealed that an analogous gene, phosphate starvation response 1 (PHR1), functions to elicit changes in root growth and morphology, anthocyanin production, and the activities of several P deficiency-inducible genes. These results suggest that the key regulators of the P deprivation responses in plants and C. reinhardtii (and perhaps other algae) are very similar, suggesting the possibility of common downstream signaling factors (42). However, it is not known whether PSR1 and PHR1 have dual functions, acting as both the sensor of intracellular Pi levels and the transcription factor that elicits altered gene expression. Recent findings have implicated an Arabidopsis SUMO E3 ligase, SIZ1, in the regulation of PHR1 activity (31), suggesting a multilevel control of plant responses to Pi deficiency.

In this study, we combine genome-based, molecular, and physiological approaches to examine P starvation responses in a wild-type C. reinhardtii strain and its isogenic psr1 mutant. Potential P deprivation-responsive genes were identified by an examination of genomic information and by cDNA-based microarray analyses. A number of new P deficiency-responsive target genes were identified, many encoding proteins that would enable the cells to better cope with excess absorbed excitation energy, limiting the extent of potential photodamage during P deprivation; these stress-related genes may be directly under PSR1 control. The importance of these genes was suggested by experiments in which elevated light levels caused a dramatic loss of viability of psr1 mutant, but not wild-type, cells during P deprivation.


Strains and growth conditions.

Chlamydomonas reinhardtii wild-type strain CC-125, a psr1-1 mutant strain that was backcrossed five times to CC-125 (54), and pKS1-31, a psr1-1 mutant strain complemented with the pKS1 plasmid containing a PSR1 genomic clone (54), were used in the experiments. For RNA preparations, cultures were grown to mid-logarithmic phase in Tris-acetate-phosphate (TAP) medium on a rotary shaker (~150 rpm) at 25°C, with continuous illumination (~40 μmol photons m−2 s−1). P-free Tris-acetate (TA) medium was prepared by substituting 1.5 mM potassium chloride for potassium phosphate, as described by Quisel et al. (39). To starve cells for P, cells were initially grown in TAP medium, harvested by centrifugation (4,000 × g, 5 min), washed twice with TA medium, and resuspended to a final cell density of 1 × 106 cell/ml in TA medium. P-starved cultures were maintained under the same growth conditions as the unstarved cells, except that all cultures were sampled 4, 12, 24, and 48 h after the initiation of P deprivation. For RNA preparation, cells were harvested by centrifugation in a model HN-SII clinical centrifuge (International Equipment Co.) at three-fourths the maximum speed (~5,800 rpm) for 10 min, and cell pellets were immediately frozen in liquid N2 and stored at −80°C. Strains for high-light-intensity experiments were grown in liquid and solid TA and TAP media and in TA medium supplemented with 10 μM glucose-1-phosphate (Sigma Chemical Co., St. Louis, MO) at 27°C, with continuous illumination at 600 to 800 μmol photons m−2 s−1.

RNA preparation.

Cell pellets were maintained in liquid N2. Each pellet was simultaneously thawed and homogenized in 8 ml of homemade Trizol reagent (Invitrogen Co., Carlsbad, CA) by continuously vortexing the suspension; the homogeneous suspension was then incubated at room temperature for at least 10 min. Chloroform (1.6 ml) was added to the suspension, and the tubes were shaken vigorously for 1 min and incubated at room temperature for an additional 2 to 3 min. Phase separation was achieved by centrifugation at 12,000 × g for 15 min, and 0.5 volume (3 to 4 ml) of 0.8 M sodium citrate-1.2 M NaCl was added to the aqueous phase, followed by the addition of 0.5 volume (compared to the initial volume) of isopropanol to precipitate nucleic acids. Precipitations were done at room temperature for 10 min, and the nucleic acid precipitate was collected by centrifugation at 12,000 × g for 10 min. Nucleic acids were washed with 8 ml of 75% ethanol, dried, and dissolved in diethyl pyrocarbonate-treated, double-distilled water. Poly(A) RNA was prepared using a MicroPoly(A) Pure small-scale mRNA purification kit (Ambion, Austin, TX) and was quantified fluorimetrically using a RiboGreen RNA quantification kit (Molecular Probes, Eugene OR) and a Tecan SPECTRAFluor fluorimeter (Zurich, Switzerland).

Preparation of fluorescent microarray probes.

The incorporation of Cy3-dUTP and Cy5-dUTP (Amersham Pharmacia Biotech, Piscataway, NJ) was performed as described by Zhang et al. (56), with the following modifications: (i) 200 to 500 ng of poly(A) RNA was used as the template for labeling reactions; (ii) the deoxynucleoside triphosphate mixture contained 1 mM dTTP and 2.5 mM (each) of dATP, dCTP, and dGTP; (iii) labeling reactions were incubated at 42°C for 30 min, 45°C for 20 min, and 50°C for 30 min; (iv) incorporation of the fluorescent dyes into cDNA was not examined; (v) the reference RNA was isolated from log-phase cells of wild-type strain CC-125 prior to P deprivation; and (vi) samples prepared from P-starved (4, 12, 24, and 48 h) CC-125 and P-replete (0 h) and P-starved (4, 12, 24, and 48 h) psr1-1 cells were compared to the reference sample.

Preparation, hybridization, washing, and scanning of microarrays.

The printing and storage of array slides (v1.1) were performed as described by Zhang et al. (56) for the “2.7 k array,” except that 3,079 different cDNA probes were printed on each slide. Additional details about the array are given at http://nostoc.stanford.edu/jeff/lab/chlamyarray/index.html and in Table Table11 and Table S1 in the supplemental material. Slides were hydrated by holding them above a 90°C water bath for ~3 s and were then immediately dried on a 100°C hot plate for ~5 s. DNAs were cross-linked by UV radiation at 300 mJ in a UV Stratalinker 1800 (Stratagene, La Jolla, CA). Prehybridization was performed by baking the slides at 65°C for 10 min, followed by a 30- to 60-min incubation at 50°C in prehybridization buffer containing 3× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate), 0.1% sodium dodecyl sulfate, and 0.1 mg/ml bovine serum albumin. Slides were washed with water and then isopropanol, each for 2 min at room temperature, and then dried by centrifugation for 5 min in a SpeedVac Plus, model SC210A (Savant, Holbrook, NY). Hybridization, washing, and scanning of the slides were performed as described by Zhang et al. (56), except that the hybridization solution was prepared by mixing 20 μl of hybridization buffer with 20 μl of the labeled probes, and the fluorescence images from 12 complete array sets (three slides per time point, with four copies of each cDNA per slide) were analyzed. Two independent sets of RNA samples were used for array analyses; a dye swap was performed using one of the RNA samples.

Transcripts showing ≥2.5-fold change in transcript abundance during phosphate deprivation

Analyses of microarray data.

Images from scanned slides were imported into the Genepix Pro 3.0 program (Axon Instruments, Union City, CA), where the spot positions were defined and their intensities determined. Spot signals that were distorted by dust, high local background, or printing flaws or signals with high proportions of saturated pixels were not included in subsequent analyses. The data were imported into Genespring 6.1 software (Silicon Genetics, Redwood City, CA) and normalized using the software's standard “per spot” and “per chip” intensity-dependent (Lowess) normalization (http://stat-www.berkeley.edu/users/terry/zarray/Html/normspie.html). Error models were computed based on replicates. Signal ratios were considered to meet threshold criteria if they passed Student's t test for significance with a P value of ≤0.05 and a multiple testing correction adjustment using the method of Benjamini and Hochberg (2).

Gene identification.

To identify specific genes in the C. reinhardtii genome, either sequence information from GenBank for previously isolated and characterized genes or sequence information from cDNAs or genomic DNAs from other organisms (mainly A. thaliana and cyanobacteria) was used to perform BLAST alignments against C. reinhardtii genomic (http://genome.jgi-psf.org/chlre2) and expressed sequence tag (EST) database (http://www.biology.duke.edu/chlamy_genome/search.html) sequences. Candidate orthologous and paralogous predicted proteins were aligned with each other and evaluated using CLUSTALW (48); identity shading was performed using GeneDoc software (33).

Quantitative real-time PCR.

Isolated total RNA was treated with RNase-free DNase I (Ambion Inc., Austin, TX) and then extracted with phenol-chloroform to isolate the RNA. Real-time quantitative PCRs (qPCR) for all genes except PTOX1, PTOX2, GAP1, and DHAR were performed by using 0.1 μg of DNase-treated total RNA and an iScript One-Step RT-PCR kit with SYBR green (Bio-Rad Laboratories, Hercules, CA). The amplifications were performed using either an iCycler IQ or Chromo4 real-time PCR thermocycler (Bio-Rad Laboratories, Hercules, CA), with the following cycling conditions: (i) 50°C for 30 min for cDNA synthesis, (ii) 95°C for 5 min to denature reverse transcriptase, and (iii) 40 to 42 cycles of 95°C for 15 or 30 s and 60°C for 30 s, with fluorescence detection after the 60°C annealing/extension step. Melting curve analysis was performed on all PCRs to ensure that single DNA species were amplified, and product sizes were verified by agarose gel electrophoresis. Discrete products were not obtained with the one-step system for the PTOX1, PTOX2, GAP1, and DHAR transcripts; consequently, two-step qPCR analysis was performed for these transcripts. For cDNA synthesis, 1 μg of DNase I-treated total RNA was reverse transcribed by using a Superscript II kit (Invitrogen, La Jolla, CA) as described by the manufacturer. qPCR was performed using either a DyNAmo Hot Start SYBR green qPCR kit (MJ Research Inc., Waltham, MA) or IQ SYBR green supermix (Bio-Rad Laboratories, Hercules, CA) and analyzed with the Mx3005P qPCR system (Stratagene, La Jolla, CA) or the Chromo4 system (Bio-Rad Laboratories, Hercules, CA). Cycling conditions included an initial incubation at 95°C for 10 min followed by 40 to 45 cycles of 94°C for 10 s, 55°C to 60°C for 15 s, and 72°C for 10 to 15 s. The relative expression ratio of a target gene was calculated based on the 2−ΔΔCT method (28), using the average cycle threshold (CT) calculated from duplicate measurements. Relative expression ratios from two independent experiments are reported. The CBLP gene was used as a control gene, and each primer was designed by Primer3 software to distinguish the different isozymes.


Immunological detection of PHOX.

C. reinhardtii cell cultures were concentrated by centrifugation and adjusted to a chlorophyll concentration of 1 mg/ml. The chlorophyll concentration was estimated according to the absorbance at 652 nm of 5-μl aliquots of cell suspension that were extracted in an 80% acetone-20% methanol mixture. Whole cell proteins equivalent to 10 μg of chlorophyll were solubilized in 1× Laemmli sample loading buffer, separated by discontinuous sodium dodecyl sulfate-polyacrylamide gel electrophoresis using a 7.5% polyacrylamide resolving gel, and transferred to polyvinylidene difluoride membranes (Pierce Biotechnology, Rockford, IL). Membranes were blocked for 1 h in a solution containing 5% dry milk and then incubated for at least 2 h with a 1/500 dilution of anti-PHOX antiserum (18) in 1% milk. A 1/3,000 dilution of alkaline phosphatase-conjugated goat anti-rabbit antiserum in 1% milk was used as the secondary antibody, and the alkaline phosphatase colorimetric reaction was performed as described by Sambrook et al. (44).


Microarray analyses.

We used cDNA microarrays (56) to analyze the effects of P deficiency on transcript abundance in both a wild-type strain (CC-125) and a mutant (psr1-1) of C. reinhardtii; this mutant is abnormal in its responses to P deprivation (54). The relative levels of nearly 3,000 different transcripts of CC-125 were analyzed over a time course of P starvation. RNAs were isolated from cells 0, 4, 12, 24, and 48 h after they were transferred from nutrient-replete medium to medium devoid of P. Using cDNA-based arrays, the levels of transcripts at each of the time points were compared to those of CC-125 cells grown in nutrient-replete TAP medium (0 h). Transcript levels were filtered for significant elevation or diminution at one or more of the time points following the initiation of P starvation. The significance of the changes observed was tested using Student's t test. Venn diagrams were constructed to show the numbers of genes for which the levels of transcripts changed by ≥2-fold (Fig. (Fig.1A),1A), ≥2.5-fold (Fig. (Fig.1B),1B), and ≥3-fold (Fig. (Fig.1C),1C), specifically in CC-125, in the psr1-1 mutant, or in both strains. We chose to concentrate on those transcripts that exhibited a change of 2.5-fold or more following exposure of cells to P deprivation conditions. Table Table11 shows the genes encoding the transcripts that changed by ≥2.5-fold in either wild-type cells or psr1-1 cells; to be included in the table, this change had to be observed for at least one point during the time course of P deprivation, and it had to pass Student's t test for significance. The transcripts of 235 genes exhibited changes of ≥2.5-fold in wild-type cells during P deprivation, corresponding to somewhat less than 10% of the genes on the array. Of these, transcripts of 152 genes differentially accumulated in wild-type cells, but not in the psr1-1 mutant; such genes are candidates for being directly regulated by PSR1. Furthermore, there were 29 transcripts in psr1-1 cells, but not in wild-type cells, that changed in abundance during P deprivation. The genes encoding these transcripts most likely respond to secondary stress conditions that result from the inability of psr1-1 cells to acclimate properly to P deficiency. An additional 83 transcripts differentially accumulated in both CC-125 and the psr1-1 mutant during P deprivation, with the majority (53) being regulated in the same direction (either up or down) in the two strains, suggesting that PSR1 is not involved (or has minimal involvement) in the control of these genes. There was one transcript that increased by >2.5-fold in CC-125 and decreased in the psr1-1 mutant; conversely, there were 30 transcripts that increased by 2.5-fold in the psr1-1 mutant and decreased in the wild-type strain. These responses may also be primary or secondary responses to PSR1 production in the cell.

FIG. 1.
Genes that exhibit altered transcript abundance during P deprivation. Proportional Venn diagrams representing genes in microarray experiments with altered transcript levels during P starvation are shown. The areas of the circles and of the overlapping ...

The genes in Table Table11 are categorized according to the putative functions of their protein products, and within each category, those genes with similar patterns of transcript accumulation are grouped together. The subset of genes for which transcripts increase or decrease in wild-type cells, but not in the psr1-1 strain during P starvation, may be controlled either directly or indirectly by PSR1. Among the genes in the category characterized by increased transcript abundance following P deprivation of wild-type cells are those encoding putative high-affinity Pi transporters, similar to Pho89 and Pho84 of S. cerevisiae; the C. reinhardtii genes encoding these transporters are designated PTB2, PTB4, and PTA3. The PTB2 transcript level increased 20-fold or more during P deprivation. These Pi transporters may contribute to the high-affinity Pi uptake activity that has been associated with P-starved cells (24, 46).

A second set of six genes that may be under PSR1 control encode potential “electron valves,” which are enzymes that may serve to protect the cell against oxidative damage resulting from overexcitation of the photosynthetic and respiratory electron transfer chains. Genes in this category encode AOX1, an isoform of a mitochondrial alternative oxidase, PTOX1, a terminal oxidase present in the plastid, and HYD2, an iron hydrogenase that can serve as an alternative electron acceptor for PS I. Interestingly, the transcripts for a plastid-targeted isoform of the glyceraldehyde-3-phosphate dehydrogenase (GAP; Calvin cycle enzyme) gene, GAP1, and for a starch phosphorylase gene, GPLV, also increase in a PSR1-dependent manner. Increasing carbon fixation coupled with the synthesis and storage of starch is another way that the cell can eliminate excess reductant generated by photosynthetic electron transport and diminish the probability of producing reactive oxygen species.

There are also 26 transcripts that increase in a PSR1-independent manner during P deprivation; the levels of these transcripts change to a similar extent in wild-type cells and the psr1-1 mutant following the imposition of P deprivation. A number of genes encoding proteases and chaperones, listed under the heading “proteolysis/stress-related proteins” in Table Table1,1, behave in this fashion. While we cannot formally rule out the possibility that genes which are regulated similarly in both the wild-type and psr1 mutant strains may respond to the centrifugation and washing steps that precede the induction of P deprivation, in many cases elevated levels of the transcripts from these genes were observed at earlier time points following P starvation in the psr1-1 mutant than in wild-type cells. Also, many of the responses were sustained over the 48-h time course of P deprivation. This suggests that these responses are neither induced by mechanical stress nor under PSR1 control but that the management of P resources (e.g., scavenging P from external and internal stores) in the mutant may be less effective than that in wild-type cells, which in turn can result in more rapid manifestation of PSR1-independent P deprivation responses. The levels of many transcripts encoding proteins required for photosynthesis also declined to similar extents in both strains. Overall, while a distinct subset of genes that respond to P deprivation conditions appears to be controlled by PSR1, a similar number of genes that are differentially expressed during P starvation are regulated by mechanisms that do not involve this transcription factor.

Genes potentially regulated by P availability based on genomic sequences.

To analyze the expression of C. reinhardtii genes involved in the acquisition of Pi from the environment (specific responses), we surveyed the draft genome sequence for genes encoding potential secreted phosphatase enzymes and Pi transporters. A single gene encoding a potential phosphatase and multiple genes encoding potential Pi transporters were identified. Changes in the levels of the transcripts from these genes were examined using qPCR following exposure of cells to P starvation.

(i) PHOX, a calcium-dependent alkaline phosphatase.

Using TBLASTN, we identified a gene on scaffold 86 of the C. reinhardtii version 3 genome sequence with strong similarity (E value = 0.0) to the PHOX alkaline phosphatase gene of Volvox carteri (18), as shown in the alignment in Fig. Fig.2.2. PHOX is likely to be the abundant, inducible, extracellular alkaline phosphatase of C. reinhardtii that was characterized by Quisel et al. (39). This enzyme has a molecular mass of ~190 kDa, and its activity was demonstrated to be Ca2+ dependent and to peak in the basic pH range. The gene model on scaffold 86 (fgenesh1_pg.C_scaffold_86000013) has 57% identity with the PHOX gene of V. carteri (Fig. (Fig.2).2). A sequence gap on the genome scaffold coincides with a region close to the C-terminal end of the predicted C. reinhardtii protein that is missing in the alignment with V. carteri PHOX. Furthermore, no other significant matches to PHOX were identified in the genome. These results suggest that C. reinhardtii may have a single PHOX gene homolog and that this single gene is interrupted by a sequence gap in the draft genome sequence.

FIG. 2.
Amino acid alignment of V. carteri and C. reinhardtii PHOX polypeptides. The diagram shows a CLUSTALW alignment of V. carteri PHOX (accession no. CAA10030 ...

(ii) Pi transporter type B family (Pho89 homologs).

At least eight potential homologs of the gene for the Saccharomyces cerevisiae Pho89 high-affinity Na+/Pi symporter are present in the C. reinhardtii genome (Fig. (Fig.3).3). Two additional gene models with significant similarity to the PTB genes were not included in this analysis, as the sequences used for the gene model predictions were of poor quality. Interestingly, pairs of these genes (PTB2 and PTB3, PTB9 and PTB12, and PTB4 and PTB5) are in close proximity and have the same 5′-to-3′ orientation in the genome, suggesting that they may have arisen from relatively recent gene duplications (Fig. (Fig.3A).3A). The amino acid sequence identity is 96% between PTB2 and PTB3, 97% between PTB9 and PTB12, and 96% between PTB4 and PTB5. PTB6 is more closely related to PTB4, PTB5, PTB9, and PTB12 (69 to 71% identity) than to PTB2 and PTB3 (38% identity), and PTB1 is the most diverged member of the family, containing a large insertion of 1,107 amino acids and only exhibiting 34 to 43% identity to the other PTB polypeptides (when this insertion is excluded from the alignment) (Fig. (Fig.3B).3B). This large, hydrophilic loop in PTB1 was described previously by Kobayashi et al. (24), who isolated the gene in a screen for arsenate-resistant C. reinhardtii mutants. The sequence identities of the C. reinhardtii PTB proteins to Pho89 from S. cerevisiae range from 19 to 23%, but the alignments show regions that are highly conserved between the yeast and algal proteins, especially at the N and C termini (data not shown).

FIG. 3.
Pi transporter type B paralogs. (A) Schematic representation of the arrangement of PTB genes in the C. reinhardtii genome. The direction of transcription is indicated by the direction of the arrows. Coding sequences are represented by light gray boxes ...

(iii) Pi transporter type A family (Pho84 homologs).

The C. reinhardtii genome contains genes for four potential homologs (PTA1 to PTA4) of the S. cerevisiae Pho84 high-affinity H+/Pi cotransporter (Fig. (Fig.4).4). Three of the PTA genes are on the same scaffold (Fig. (Fig.4A),4A), again suggesting that the expansion of this gene family is evolutionarily recent. The PTA3 and PTA4 proteins are the most similar, with 88% identity, but all of the PTA polypeptides are highly conserved, with a minimum of 73% sequence identity (Fig. (Fig.4B).4B). Only 20 to 21% of the amino acids are conserved between the PTAs and Pho84, but a number of highly conserved motifs are located throughout the entire length of the sequences, with the exception of the very N- and C-terminal regions (data not shown).

FIG. 4.
Pi transporter type A paralogs. (A) Schematic representing the arrangement of PTA genes in the C. reinhardtii genome. The PTA1, PTA2, and PTA3 gene models are based on alignments with the cDNA sequences (accession no. ...

Expression of PHOX, PTB, and PTA genes.

qPCR analyses were performed to confirm the changes in transcript levels observed on the microarray and also to determine whether potentially P-regulated genes identified from the genomic sequence are controlled at the level of mRNA abundance. As previously noted, microarray analyses revealed increased PTB2, PTB4, and PTA3 mRNA levels in wild-type cells, but not in the psr1 mutant, during P starvation (Table (Table1).1). For PTB2 and PTB4, peak increases in transcript abundance occurred at 24 h following the transfer of cells to medium devoid of P, with a slight decline in the level after 48 h; therefore, we considered 24 h to be an appropriate starvation time to examine transcript abundance for other members of the Pi transporter gene families. The qPCR results for transcripts encoded by PHOX and the different PTB and PTA gene family members from both wild-type cells and the psr1-1 mutant after 24 h of P starvation are presented relative to those for the same transcripts in wild-type cells or psr1-1 mutant cells in the logarithmic phase of growth under Pi-replete conditions (0-h references).

As shown in Table Table2,2, for wild-type cells (CC-125) transcripts encoding many of the predicted transporters increased after 24 h of P deprivation. Significant increases in PTB2, PTB3, PTB4, and PTB5 mRNAs were observed for wild-type cells after 24 h of P deprivation, with little increase or a declining abundance in the psr1-1 mutant subjected to the same conditions. These results demonstrate that PSR1 is involved in controlling the abundance of these transcripts and that this control is probably exerted at the level of transcription (since PSR1 is a MYB domain-containing transcription factor). Interestingly, PTB2-PTB3 and PTB5-PTB4 are contiguous, with the transcript from the proximal gene of each pair (PTB2 and PTB5) being most strongly elevated after 24 h of P starvation. No significant increase in the level of PTB1 mRNA was observed during P stress, which is in agreement with the report of Kobayashi et al. (24), and the PTB6 transcript could not be detected (data not shown).

Putative Pi scavenging-related transcript abundance after 24 h of Pi deprivation

In contrast, expression patterns of the PTA transcripts varied considerably during P stress (Table (Table2).2). In wild-type cells, PTA1 transcript abundance was reduced 1,000-fold during P deprivation but only 2- to 6-fold in the psr1 mutant (depending on whether the 24-h time point is compared to the wild-type or psr1-1 0-h control), suggesting that PSR1 may play a role in the repression of this gene. PTA2 and PTA3 mRNA levels declined slightly or remained unchanged after P deprivation, whereas PTA4 expression increased significantly in both the wild-type and psr1 mutant strains. Interestingly, the level of the PTA4 transcript was 10- to 20-fold higher in P-starved wild-type cells than in P-starved psr1 cells. It is possible that PSR1 is partially responsible for the induction of this gene, or alternatively, the psr1 mutant may be unable to fully activate the expression of PTA4 due to secondary stress responses that occur in the mutant following the imposition of P deprivation. In sum, the results suggest that the PTA transporters (except, perhaps, for PTA4) are likely to not contribute significantly to the high-affinity Pi transport activity observed soon after the imposition of P deprivation. These transporters could be involved in transport under nutrient-replete conditions or may modulate Pi transport between specific cellular compartments. Alternatively, it is possible that some of the PTA proteins transport solutes other than Pi.

The qPCR analysis demonstrated that PHOX transcript levels markedly increase by 24 h of P depletion (Table (Table2).2). This increase is under the control of PSR1, and the extent of the increase may reflect the very low basal level of the transcript present in P-replete cells. To confirm that the C. reinhardtii PHOX gene encodes the major inducible alkaline phosphatase activity, immunoblot analysis was performed on whole-cell extracts from Pi-replete versus P-starved wild-type and psr1-1 mutant strains with antiserum raised to a V. carteri PHOX peptide (Fig. (Fig.5)5) (18). The immunoblot revealed a polypeptide of ~190 kDa that cross-reacts with anti-PHOX antibodies; this polypeptide was observed in extracts of wild-type cells starved for P (but not in extracts of cells grown on complete medium), and the band was not detected for the psr1-1 strain under any conditions tested. The cross-reacting band was not detected on a control blot without the anti-PHOX antibody (data not shown), excluding the possibility that residual alkaline phosphatase activity remained in the C. reinhardtii protein extract after the proteins were transferred to the membrane (since an alkaline phosphatase color reaction was used to detect the secondary antibody). Together, these data strongly suggest that C. reinhardtii PHOX encodes the major inducible alkaline phosphatase activity of C. reinhardtii.

FIG. 5.
Immunodetection of PHOX. The immunoblot was created by using anti-PHOX antiserum (18) and whole-cell extracts from either P-replete (+) or P-deprived (24 h) (−) wild-type and psr1 mutant cells.

Expression of genes encoding potential “electron valves.”

The psr1 mutants were originally isolated based on their inability to either perform high-affinity Pi uptake or synthesize extracellular phosphatase activity following exposure of C. reinhardtii to P starvation conditions (46, 54). The psr1-1 and psr1-2 (two different alleles) mutants were able to down-regulate photosynthetic electron transport and did not show apparent signs of photodamage or reduced viability relative to wild-type cells, even after 10 days of growth without P. This is in contrast to the phenotype of the sac1 strain, which is unable to acclimate to sulfur deprivation; this mutant is very sensitive to light conditions, and in the absence of sulfate, it dies under moderate light intensities (50 to 100 μmol photons m−2 s−1) (10). As a consequence, PSR1 was originally considered to regulate specific and not general P deprivation responses. However, as noted above, microarray analyses revealed that a number of transcripts encoding proteins associated with conditions of hyperstimulation of photosynthetic electron transport increased and that these increases appeared to be under PSR1 control. qPCR was used to establish a more quantitative assessment of the levels of these transcripts during P deprivation (Table (Table3).3). The AOX1 transcript increased three- to fivefold in wild-type cells starved for P for 24 h relative to unstarved cells. P deprivation had little effect on the level of this transcript in the psr1-1 strain, corroborating the results of the microarray experiments (Table (Table1).1). In contrast, the AOX2 transcript was not detected under any cell growth conditions used (data not shown). The transcripts from both genes encoding the plastid terminal oxidases (PTOX1 and PTOX2) (25, 38) were detected by qPCR, but in contrast to the microarray results, where a 13-fold upregulation of PTOX1 mRNA was observed in wild-type cells after 48 h of P starvation (Table (Table1),1), PTOX1 transcripts appeared to increase only 2.5- to 3-fold, while the level of the PTOX2 transcript was not consistently higher in wild-type cells starved for P than in unstarved cells. A small change in PTOX2 transcript levels was also observed in the psr1-1 mutant. The discrepancies between the microarray and qPCR results may be a consequence of the fact that the qPCR is much more sensitive than microarray analyses. For example, the fluorescence signal of PTOX1 in the microarray experiments was comparable to the background at most time points. Consequently, even small changes in the abundance of the transcript registered as large changes in the observed ratios. HYD1 and HYD2 represent another pair of paralogous genes that appeared to be differentially expressed during P deprivation. Consistent with the microarray results, qPCR measurements showed that the HYD1 mRNA did not change over the course of P deprivation in wild-type cells. In contrast, HYD2 mRNA abundance increased 8- to 13-fold in P-starved wild-type cells. These observations support the inference drawn from the microarray results that PSR1 may directly or indirectly influence the expression of the HYD2 gene. The abundance of both HYD1 and HYD2 declined severely (~5- to 10-fold) in the psr1-1 strain, possibly reflecting some contribution of PSR1 to the maintenance of basal levels of HYD expression. PSR1-dependent regulation was also observed for two genes encoding the starch phosphorylase homologs GPLV1 and GPLV2. Based on both microarray and qPCR data, the GPLV1 transcript was elevated four- to ninefold in wild-type cells after 24 h of P deprivation but remained approximately constant in the psr1-1 mutant during P deprivation. Similarly, the GPLV2 transcript increased threefold in wild-type cells and remained constant in the psr1-1 strain during P deprivation. Therefore, in most of the cases that we have examined, where transcript abundance for one member of a gene family is potentially regulated by PSR1, the regulation of the other is independent of PSR1.

“Electron valve”-encoding transcript abundance after 24 h of Pi deprivation

Three genes encoding glyceraldehyde-3-phosphate dehydrogenases (GAPs) have been identified in the C. reinhardtii genome. The GAP1 and GAP3 polypeptides have N-terminal extensions. GAP3 has been characterized biochemically and encodes a subunit of the chloroplastic NADP-dependent GAP that functions in the Calvin-Benson cycle (27). Although GAP1 is predicted to be targeted to the mitochondrion, the prediction has a low confidence score (TargetP score, 0.635), and generally for C. reinhardtii, prediction programs are not reliable in discriminating between mitochondrial and chloroplast targeting (12). GAP2 has no presequence and is likely to encode the cytosolic form of the enzyme that functions in glycolysis. GAP1 is the only member of this gene family that is represented on the microarray, and the level of its transcript appeared to increase five- to sixfold during P starvation in wild-type cells, but not in the psr1-1 mutant (Table (Table1).1). Based on qPCR, the GAP1 mRNA increased 37- to 72-fold in wild-type cells after 24 h of P deprivation. The qPCR data suggest that the GAP2 mRNA increases approximately four- to sixfold in wild-type cells, but not in psr1-1 mutant cells. In contrast, the GAP3 RNA abundance is relatively constant in both wild-type cells and the psr1-1 mutant (Table (Table3).3). These results suggest that GAP1 and GAP2, but not GAP3, are under the control of PSR1 and also raise the possibility that PSR1 influences the expression of glycolytic enzymes. Interestingly, although some plants express a nonphosphorylating GAPN enzyme to “bypass” the NAD+- and Pi-dependent enzyme in glycolysis (11, 17) and although a GAPN gene homolog is present in the C. reinhardtii genome, the mRNA from this gene declined 5- to 10-fold during P deprivation, independent of PSR1 (data not shown). These results suggest that C. reinhardtii uses mechanisms to regulate carbon flux through the glycolytic pathway that are different from those described for vascular plants.

Sensitivity of psr1-1 mutant to high light intensity.

The discovery that the mRNAs of a variety of genes with a putative electron valve function increased during P deprivation in a PSR1-dependent manner raised questions concerning the role that PSR1 might play in controlling general responses to P deficiency. Therefore, we compared the sensitivities of the psr1-1 mutant, wild-type cells, and a PSR1-complemented strain to elevated light intensities during P deprivation (Fig. (Fig.6).6). In moderate light (40 μmol photons m−2 s−1) on TA medium supplemented with a limiting amount of a hydrolyzable Pi source (10 μM glucose-1-phosphate), the psr1-1 mutant survived but grew more slowly than wild-type cells or the complemented strain and also arrested at a lower cell density. In contrast, when the three strains were exposed to high light intensities (700 to 800 μmol photons m−2 s−1) for 3 days during P deprivation, the psr1-1 mutant became bleached and died, while little loss of viability occurred for wild-type cells and the complemented mutant (Fig. (Fig.6A).6A). Viability staining of cultures grown in liquid TA medium under high light intensities confirmed that 80 to 90% of psr1-1 cells died by day 6 of P depletion, whereas 80% of the wild-type cells and the complemented cells survived these conditions (Fig. (Fig.6B6B).

FIG. 6.FIG. 6.
Light sensitivity of psr1 during P deprivation. (A) Comparison of growth of wild-type cells, psr1 mutant cells, and a psr1-complemented strain on solid TAP medium (left panels) or TA medium supplemented with 10 μM glucose-1-phosphate (right panels). ...


We have used genomic information to help elucidate the effects of Pi starvation on gene expression in C. reinhardtii. Microarray studies were used to identify genes that are either positively or negatively regulated as cells become starved for P and to define those genes that appear to be controlled by the transcription factor PSR1 (46, 54). Similar microarray analyses of P deficiency responses have been performed with Arabidopsis thaliana (19, 52). A few qualitative similarities are observed between the responses in C. reinhardtii and those in plants, including induction of genes encoding P-scavenging enzymes and down-regulation of photosynthetic genes and genes involved in cytoplasmic and chloroplast translation, along with modulation of some genes involved in carbon metabolism (52). Overall, the overlap between P deficiency-responsive target genes in C. reinhardtii and those in plants appears to be limited, perhaps reflecting differences in P storage and distribution between the multicellular plants and the unicellular algae. It should also be considered that all of the array experiments in question were performed with subgenome arrays and that some common target genes may not have been represented on the plant or C. reinhardtii arrays.

In initial studies of psr1 mutants, no increased sensitivity to moderate light intensities was observed, and accordingly, PSR1 was not considered to play a role in protecting cells against photodamage. However, the microarray data demonstrated that a class of genes encoding potential “electron valves” appeared to be activated during P starvation in a PSR1-dependent manner. These genes encode enzymes that assist in the dissipation of potentially harmful species that may accumulate as a consequence of photosynthetic and respiratory electron transfer reactions under conditions that might restrict the use of products generated by these reactions. Microarray analyses identified transcripts from 210 genes that exhibited differential accumulation during P deprivation of threefold or greater in the wild-type and psr1-1 strains (Fig. (Fig.1).1). This represents approximately two-thirds the number of transcripts that exhibited a threefold change during sulfur starvation (56), although some transcripts are common between the two sets of data. The commonality may reflect responses to general intracellular stress conditions that may be generated by both types of nutrient deprivation.

Pi-scavenging genes.

A total of 10 potential Pi transporters were identified in the C. reinhardtii genome (Fig. (Fig.33 and and4);4); at least six are differentially expressed during P deprivation. The transcripts for four of the PTB and one of the PTA genes increased in abundance in P-starved cells. Increased levels of these transcripts appeared to require PSR1. It is not clear which of these transporters (or groups of transporters) contributes most significantly to the high-affinity Pi uptake observed in P-deprived cells. In S. cerevisiae, which grows best in low-pH environments, the major secreted phosphatase, PHO5, has an acidic pH optimum, and the dominant high-affinity transporter is the Pho84 H+/Pi symporter (PTA type) (3, 6). The S. cerevisiae Pho89 Na+/Pi cotransporter (PTB type) may play a less significant role in Pi uptake, as it operates best in alkaline conditions (29). However, the situation may be very different for C. reinhardtii, which thrives in neutral or basic environments and secretes a predominant alkaline phosphatase during P deprivation (39). This may, in part, explain the bias towards activation of the PTB genes in C. reinhardtii; like Pho89, the PTB enzymes would be expected to display superior Pi uptake at an alkaline pH. Currently, there is no definitive evidence regarding the cellular location of the PTB and PTA proteins, but TargetP v1.01 predicts with high confidence that all of the PTB proteins are routed along the secretory pathway (Table (Table4),4), making these strong candidates for components of a plasma membrane (or vacuolar membrane)-associated Pi starvation-inducible Pi uptake system. Based on both microarray and qPCR analyses, PTB2 and PTB5 transcripts increase in abundance >20- and several hundredfold, respectively, following exposure of cells to P deprivation conditions (Tables (Tables11 and and2).2). No change in Pi uptake characteristics was observed with a ptb1 mutant of C. reinhardtii (15), implying that the gene product is not essential for the elevated Pi uptake observed when cells begin to experience P limitations. Congruent with this result, the PTB1 transcript level does not respond significantly to P deprivation conditions. Kobayashi et al. (24) have suggested that PTB1 may be involved in intracellular Pi transport. The paired genomic clusters of the PTB2/PTB3 and PTB4/PTB5 genes, the high sequence identity between the members of the pairs, and the presence of numerous, scattered PTB gene fragments in close proximity suggest that this gene family has undergone a recent expansion. The different patterns of transcript abundance in response to P limitation defined for the individual members of this gene family suggest that the encoded proteins may play divergent roles in Pi uptake and distribution within the cell.

TargetP prediction of subcellular targeting of potential Pi transporters

The family of transporters designated PTA (Pho84 homologs) for C. reinhardtii has been designated Pht for A. thaliana. These proteins in A. thaliana have been implicated in both high- and low-affinity Pi transport, and the genes encoding these proteins are specifically expressed in plant roots (47). For C. reinhardtii, only PTA2 is predicted to be routed through the secretory pathway, and this prediction has a low confidence score (Table (Table4).4). Furthermore, changes in transcript levels during P deprivation were varied for the different PTA genes. There was a strong PSR1-dependent decline in PTA1 mRNA and no significant change in the level of the PTA2 transcript. Microarray experiments suggested a PSR1-dependent increase in PTA3 mRNA levels, but this observation was not confirmed by qPCR measurements (Tables (Tables11 and and2).2). In cases where the qPCR and microarray data do not corroborate each other, we generally find that the fluorescence signal of the corresponding gene on the microarray is low (comparable to the background), and this can result in faulty ratios. The PTA4 transcript increased significantly in both wild-type and psr1-1 mutant cells exposed to P deprivation, although the increase was consistently higher for wild-type cells. We have previously demonstrated that the activation of high-affinity Pi uptake in P-deficient cells requires PSR1 (46). Therefore, in general, the patterns of PTA transcript accumulation in wild-type and mutant cells (except, perhaps, for PTA4) and the predicted subcellular locations of the PTA polypeptides are not consistent with a role for these transporters in P starvation-inducible Pi uptake. It is more likely that the PTA gene products contribute to either low-affinity Pi uptake or intracellular trafficking of Pi. For example, PTA1 might encode a component of a low-affinity Pi uptake system that is expressed under nutrient-replete conditions and repressed when the cells are starved for P. A more detailed biochemical approach needs to be initiated in order to develop a precise understanding of the functions of the different PTA and PTB transport proteins.

The C. reinhardtii PHOX gene encodes a homolog of the PHOX protein of V. carteri, and the level of mRNA encoding this protein increases dramatically during P deprivation (Table (Table2).2). Immunological analyses using V. carteri PHOX antibodies suggest that this protein represents the previously characterized secreted, Ca2+-dependent alkaline phosphatase of C. reinhardtii (39).

C. reinhardtii also maintains internal stores of polyphosphate in cytosolic organelles, akin to the polyphosphate bodies of fungi and the acidocalciosomes found in trypanosomatids and apicomplexan parasites (43). Although nothing is known about the synthesis and regulation of polyphosphate in C. reinhardtii, it is interesting that the mRNA encoding VCX1, a potential vacuolar H+/Ca2+ antiporter (Table (Table1),1), also increased during P deprivation. In yeast, this protein is responsible for controlling cytosolic Ca2+ levels in the presence of high extracellular concentrations of Ca2+ (30). This finding raises the possibility that the VCX1 protein serves to maintain vacuolar sequestration of the Ca2+ that is released from polyphosphate bodies in the vacuole as Pi is enzymatically generated during P deprivation-triggered polyphosphate degradation.

Stress-related gene expression.

A number of transcripts encoding stress-related proteins increased during P starvation in both the wild-type strain and the psr1-1 mutant; the levels of these transcripts do not appear to be under PSR1 control. These transcripts include those encoding two putative 22-kDa chloroplast-localized heat shock proteins, a glutathione S-transferase, an E3 ubiquitin ligase, and two protease homologs (Table (Table1,1, proteolysis/stress-related proteins). Chaperones and proteases may be required to sequester and eliminate misfolded or damaged proteins (9, 13). While most of these transcripts only significantly increased at the final time point following elimination of P from the growth medium (48 h) in the wild-type strain, they became elevated by 12 h of P deprivation in the psr1-1 mutant. These results suggest that while both strains show a generalized stress response, the mutant senses more extreme stress conditions at earlier times after the elimination of P from the medium; this more rapid response may reflect the inability of the mutant to scavenge Pi from both intracellular and extracellular resources.

A number of physiological responses have been shown to be elicited by P starvation. With respect to photosynthetic activity, there is a depletion of PS II reaction centers, accumulation of QB nonreducing centers, increased nonphotochemical quenching, and transition of the energy transfer characteristics of the antenna chlorophyll from state I to state II. These characteristics are indicators of excess excitation of the photosynthetic apparatus and suggest that the photosynthetic electron transport chain is hyperreduced, even under conditions of moderate light intensity (53). Gene expression analysis supports this idea, since, in addition to the P starvation-responsive chaperones and proteases, we observed increases in some transcripts that were also elevated at high light intensities (21). For example, there was an increase in transcripts encoding the granule-bound starch synthase I (STA2), which may help to relieve excess excitation by stimulating starch synthesis, and LHCSR2, a protein related to light-harvesting complex (LHC) antenna proteins that may have a photoprotective function (Table (Table1)1) (21, 41). The transcripts for these proteins are also elevated during sulfur deprivation (56). Although the kinetics of increase in STA2 mRNA were similar for wild-type cells and the psr1-1 mutant, the increase in transcript abundance was observed earlier in psr1-1 mutant cells than in wild-type cells (12 h versus 24 h). Furthermore, the increased transcript levels were sustained in the mutant cells, even up to 48 h; at 48 h, the mRNA level in wild-type cells dropped to below the 2.5-fold threshold (Table (Table1).1). These results again suggest that the signal transduction pathway that regulates some of the deprivation-induced genes becomes active more rapidly and that the response is sustained longer following the initiation of P starvation of psr1-1 than that of wild-type cells.

Electron valves and radical-scavenging enzymes.

In addition to potential stress-related transcripts that increase in both wild-type cells and psr1-1 mutant cells, another set of transcripts encoding proteins implicated in photoprotection is elevated exclusively in wild-type cells during P deprivation. These photoprotective proteins may act as “electron valves” that serve to drain electrons from the photosynthetic and mitochondrial electron transfer chains, decreasing the potential dangers associated with hyperreduction of electron carriers and the generation of triplet chlorophyll molecules. An enhanced capacity for alternative respiration in iron-starved C. reinhardtii has been reported previously (51), and mutants of tobacco plants with a defective alternative oxidase were unable to sustain normal respiratory rates upon P deprivation (37). We observed increased transcript levels from the alternative oxidase gene AOX1 in wild-type cells starved for P for 24 h. In contrast, the psr1-1 mutant exhibited a transient increase in AOX1 mRNA after 4 h of P deprivation, but this increase was not maintained. Increased alternative oxidase activity during P starvation may perform a number of functions, including preventing the accumulation of reactive oxygen species in the mitochondria, allowing continued flux of metabolites through the tricarboxylic acid cycle for generating organic carbon “backbones,” and controlling carotenoid biosynthesis and the rate of ATP formation under conditions in which the availability of Pi may be limiting (7, 23, 25). Increased levels of the AOX1 transcript were previously reported during S starvation (56), and upon transfer of cells from growth on ammonium to growth on nitrate (1). Surprisingly, AOX1 is the only transcript encoding a known mitochondrial protein that was significantly upregulated in these experiments, suggesting that the control of mitochondrial processes during P deprivation may be regulated largely by posttranscriptional mechanisms.

Increased levels of transcripts encoding PTOX1, a putative plastid terminal oxidase, were also observed in wild-type P-starved cells, but not in the psr1-1 mutant. The PTOX1 protein may represent the terminal oxidase activity responsible for chlororespiration (8, 38). This terminal oxidase may relieve redox pressure associated with photosynthetic electron transfer by withdrawing electrons from plastoquinol and combining them with O2. PTOX1 is a homolog of a plastid terminal oxidase in plants, encoded by the IMMUTANS gene, which is also implicated in carotenoid synthesis (23), raising the possibility that PTOX1 induction during P starvation in C. reinhardtii might play a role in the production of photoprotective carotenoids.

A third potential electron valve that may increase during P deprivation in cells with active PSR1 is the Fe hydrogenase, encoded by HYD2 (HYDB). The C. reinhardtii hydrogenases catalyze the transfer of electrons from PS I to water to produce H2 gas under anaerobic/microaerobic conditions, and they are believed to help reoxidize the plastoquinol pool when the cytoplasm of the cell is highly reducing, especially during anaerobic or hypoxic growth (14, 16, 20). The hydrogenase activity might increase plastoquinol oxidation during P starvation, relieving the block on linear electron transfer and helping to prevent the production of reactive oxygen species. The activation of HYD2 gene expression might be a by-product of the production of a microaerobic environment within cells as oxygen evolution declines during P deprivation and respiration consumes oxygen faster than it is being produced. However, were this the case, it would be expected that HYD1 expression would be induced as well, but this was not observed in either microarray or qPCR experiments. HYD2 mRNA abundance is much greater in wild-type cells than in the psr1-1 mutant (compare transcript levels in Tables Tables11 and and3),3), suggesting that PSR1 plays a role in maintaining expression of this gene (or the conditions required for its expression).

As discussed above, C. reinhardtii shows increased STA2 mRNA expression during P deprivation; the STA2 protein is involved in the synthesis of starch. Elevating starch synthesis can help control the accumulation of excess reductant in the cell and limit the production of reactive oxygen species. However, P starvation of wild-type cells also leads to elevated mRNA for both GPLV, encoding a starch phosphorylase, and an alpha amylase gene. Both of these gene products are involved in the conversion of starch to sugar. Therefore, the starved cells may also have an increased capacity to actively degrade storage starch, which could be important for maintaining the flux through the starch synthesis pathway. Finally, the transcript for dehydroascorbate reductase also increases in P-starved wild-type cells but not in the psr1-1 mutant. This enzyme is involved in recycling of ascorbate, an antioxidant that plays an important role in scavenging reactive oxygen species in the chloroplast. Together, the results presented above demonstrate that increases in mRNAs for proteins that have an electron valve function and that help the cells manage an elevated redox state appear to be normal aspects of the P starvation responses of C. reinhardtii. Although a causative relationship between the activation of these pathways and resistance to high light intensities remains to be established, it is intriguing that conditions of high light intensity are lethal to the psr1 mutant and that the transcripts from many of these genes do not significantly increase in the psr1-1 mutant.


A number of genes represented on the array encoding proteins with photosynthetic functions appear to be differentially expressed during P starvation. While down-regulation of PS II function and abundance has been demonstrated previously (53), we did not observe changes in mRNA levels for nuclear genes encoding structural subunits of PS II during P starvation. Instead, transcripts from several LHCA genes, encoding components of the LHC of PS I (LHC I), and PSAE and PSAK genes, encoding subunits of PS I, declined in both P-starved wild-type cells and psr1-1 mutant cells. A reduced level of the CHL27 (CRD1) transcript was also observed. CHL27 encodes a component of the chlorophyll biosynthetic enzyme Mg-protoporphyrin IX monomethylester cyclase (49) and has been reported to play a role in controlling the interaction between the PS I reaction center and LHC I. Therefore, reduced levels of CHL27 protein would be expected to result in a weaker energetic connection between the LHC I antennae and core PS I (32). The transcripts for proteins that are minor subunits of the cytochrome b6/f complex, i.e., PETO and PETN, also appear to decline in wild-type cells, with the PETM and PETC transcripts showing similar trends (although the level of the PETM and PETC mRNAs only declined by 50% over the time course [see Table S1 in the supplemental material]). Indeed, transcripts encoding all PSB, PSA, and PET proteins represented by elements on the array declined in both wild-type cells and the psr1-1 strain.

Aberrant gene expression in the psr1-1 mutant.

Twenty-nine transcripts exhibited increased accumulation in psr1-1 but not wild-type cells during P starvation (Fig. (Fig.1).1). The increased levels of these transcripts may be a consequence of the more extreme stress conditions experienced by the psr1-1 mutant than by wild-type cells since the mutant would not be able to properly acclimate as P levels in the medium declined. The aberrant metabolic status of the mutant strain may also trigger specific signaling pathways in addition to more general stress response pathways. Increased levels of transcripts encoding phytoene synthase, a key regulatory enzyme in carotenoid biosynthesis, and homogentisic acid geranylgeranyl transferase, an enzyme required for tocopherol biosynthesis, may reflect the cell's response to elevated reducing conditions and a greater tendency to form reactive oxygen molecules, both of which are associated with nutrient deprivation conditions (34). Interestingly, the psr1-1 mutant also exhibited increased mRNAs for a subset of genes associated with sulfur acquisition and assimilation (Table (Table1,1, sulfur metabolism). This finding may reflect a change in the energy status of the cell and the inability of the mutant to efficiently take up and activate SO42−. However, since we observed an increase in transcript levels for only a subset of the genes associated with sulfur-deficient conditions, it is possible that there is a more specific link between the P and S deprivation pathways. Understanding these links may help us to decipher the networks of interactions among stress responses and the hierarchy of responses observed.

Supplementary Material

[Supplemental material]


J.L.M. was supported by a Life Sciences Research Foundation fellowship from the U.S. Department of Energy, and this work was supported by U.S. Department of Agriculture grant 2002-35301-12178 and National Science Foundation grant MCB 0235878, both awarded to A.R.G.

We thank Chung-Soon Im, Wirulda Pootakham, Steve Pollock, Nakako Shibagaki, and Jeff Shrager for technical assistance and helpful comments on the manuscript and Armin Hallmann and Sabeeha Merchant for the kind gift of antibodies. We also thank the Joint Genome Institute for providing access to a prerelease of version 3 of the Chlamydomonas draft genome sequence.


Supplemental material for this article may be found at http://ec.asm.org/.

This is Carnegie Institution publication no. 1725.


1. Baurain, D., M. Dinant, N. Coosemans, and R. F. Matagne. 2003. Regulation of the alternative oxidase Aox1 gene in Chlamydomonas reinhardtii. Role of the nitrogen source on the expression of a reporter gene under the control of the Aox1 promoter. Plant Physiol. 131:1418-1430. [PMC free article] [PubMed]
2. Benjamini, Y., and Y. Hochberg. 1995. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B 57:289-300.
3. Berhe, A., U. Fristedt, and B. L. Persson. 1995. Expression and purification of the high-affinity phosphate transporter of Saccharomyces cerevisiae. Eur. J. Biochem. 227:566-572. [PubMed]
4. Bieleski, R. L. 1973. Phosphate pools, phosphate transport, and phosphorus availability. Annu. Rev. Plant Physiol. Plant Mol. Biol. 24:225-252.
5. Brooks, A.. 1996. Effects of phosphorus nutrition on ribulose-1,5-bisphosphate carboxylase activation, photosynthetic quantum yield and amounts of some Calvin cycle metabolites in spinach leaves. Aust. J. Plant Physiol. 13:221-237.
6. Bun-ya, M., M. Nishimura, S. Harashima, and Y. Oshima. 1991. The PHO84 gene of Saccharomyces cerevisiae encodes an inorganic phosphate transporter. Mol. Cell, Biol. 11:3229-3238. [PMC free article] [PubMed]
7. Carol, P., and M. Kuntz. 2001. A plastid terminal oxidase comes to light: implications for carotenoid biosynthesis and chlororespiration. Trends Plant Sci. 6:31-36. [PubMed]
8. Cournac, L., G. Latouche, Z. Cerovic, K. Redding, J. Ravenel, and G. Peltier. 2002. In vivo interactions between photosynthesis, mitorespiration, and chlororespiration in Chlamydomonas reinhardtii. Plant Physiol. 129:1921-1928. [PMC free article] [PubMed]
9. Craig, E. A., H. C. Eisenman, and H. A. Hundley. 2003. Ribosome-tethered molecular chaperones: the first line of defense against protein misfolding? Curr. Opin. Microbiol. 6:157-162. [PubMed]
10. Davies, J. P., F. H. Yildiz, and A. Grossman. 1996. Sac1, a putative regulator that is critical for survival of Chlamydomonas reinhardtii during sulfur deprivation. EMBO J. 15:2150-2159. [PMC free article] [PubMed]
11. Duff, S. M. G., G. B. G. Moorhead, D. D. Lefebvre, and W. C. Plaxton. 1989. Phosphate starvation inducible “bypasses” of adenylate and phosphate dependent glycolytic enzymes in Brassica nigra suspension cells. Plant Physiol. 90:1275-1278. [PMC free article] [PubMed]
12. Emanuelsson, O., H. Nielsen, S. Brunak, and G. von Heijne. 2000. Predicting subcellular localization of proteins based on their N-terminal amino acid sequence. J. Mol. Biol. 300:1005-1016. [PubMed]
13. Fink, A. L. 1999. Chaperone-mediated protein folding. Physiol. Rev. 79:425-449. [PubMed]
14. Forestier, M., P. King, L. Zhang, M. Posewitz, S. Schwarzer, T. Happe, M. L. Ghirardi, and M. Seibert. 2003. Expression of two [Fe]-hydrogenases in Chlamydomonas reinhardtii under anaerobic conditions. Eur. J. Biochem. 270:2750-2758. [PubMed]
15. Fujiwara, S., I. Kobayashi, S. Hoshino, T. Kaise, K. Shimogawara, H. Usuda, and M. Tsuzuki. 2000. Isolation and characterization of arsenate-sensitive and resistant mutants of Chlamydomonas reinhardtii. Plant Cell Physiol. 41:77-83. [PubMed]
16. Ghirardi, M. L., L. Zhang, J. W. Lee, T. Flynn, M. Seibert, E. Greenbaum, and A. Melis. 2000. Microalgae: a green source of renewable H2. Trends Biotechnol. 18:506-511. [PubMed]
17. Habenicht, A., U. Hellman, and R. Cerff. 1994. Non-phosphorylating GAPDH of higher plants is a member of the aldehyde dehydrogenase superfamily with no sequence homology to phosphorylating GAPDH. J. Mol. Biol. 237:165-171. [PubMed]
18. Hallmann, A. 1999. Enzymes in the extracellular matrix of Volvox: an inducible, calcium-dependent phosphatase with a modular composition. J. Biol. Chem. 274:1691-1697. [PubMed]
19. Hammond, J. P., M. J. Bennett, H. C. Bowen, M. R. Broadley, D. C. Eastwood, S. T. May, C. Rahn, R. Swarup, K. E. Woolaway, and P. J. White. 2003. Changes in gene expression in Arabidopsis shoots during phosphate starvation and the potential for developing smart plants. Plant Physiol. 132:578-596. [PMC free article] [PubMed]
20. Happe, T., and A. Kaminski. 2002. Differential regulation of the Fe-hydrogenase during anaerobic adaptation in the green alga Chlamydomonas reinhardtii. Eur. J. Biochem. 269:1022-1032. [PubMed]
21. Im, C. S., Z. Zhang, J. Shrager, C. W. Chang, and A. R. Grossman. 2003. Analysis of light and CO2 regulation in Chlamydomonas reinhardtii using genome-wide approaches. Photosyn. Res. 75:111-125. [PubMed]
22. Jacob, J., and D. W. Lawlor. 1993. In vivo photosynthetic electron transport does not limit photosynthetic capacity in phosphate-deficient sunflower and maize leaves. Plant Cell Environ. 16:785-795.
23. Josse, E. M., A. J. Simkin, J. Gaffe, A. M. Laboure, M. Kuntz, and P. Carol. 2000. A plastid terminal oxidase associated with carotenoid desaturation during chromoplast differentiation. Plant Physiol. 123:1427-1436. [PMC free article] [PubMed]
24. Kobayashi, I., S. Fujiwara, K. Shimogawara, T. Kaise, H. Usuda, and M. Tsuzuki. 2003. Insertional mutagenesis in a homologue of a Pi transporter gene confers arsenate resistance on Chlamydomonas. Plant Cell Physiol. 44:597-606. [PubMed]
25. Kuntz, M. 2004. Plastid terminal oxidase and its biological significance. Planta 218:896-899. [PubMed]
26. Lenburg, M. E., and E. K. O'Shea. 1996. Signaling phosphate starvation. Trends Biochem. Sci. 21:383-387. [PubMed]
27. Li, A. D., and L. E. Anderson. 1997. Expression and characterization of pea chloroplastic glyceraldehyde-3-phosphate dehydrogenase composed of only the B-subunit. Plant Physiol. 115:1201-1209. [PMC free article] [PubMed]
28. Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2−ΔΔCT method. Methods 25:402-408. [PubMed]
29. Martinez, P., and B. L. Persson. 1998. Identification, cloning and characterization of a derepressible Na+-coupled phosphate transporter in Saccharomyces cerevisiae. Mol. Gen. Genet. 258:628-638. [PubMed]
30. Miseta, A., R. Kellermayer, D. P. Aiello, L. Fu, and D. M. Bedwell. 1999. The vacuolar Ca2+/H+ exchanger Vcx1p/Hum1p tightly controls cytosolic Ca2+ levels in S. cerevisiae. FEBS Lett. 451:132-136. [PubMed]
31. Miura, K., A. Rus, A. Sharkhuu, S. Yokoi, A. S. Karthikeyan, K. G. Raghothama, D. Baek, Y. D. Koo, J. B. Jin, R. A. Bressan, D.-J. Yun, and P. M. Hasegawa. 2005. The Arabidopsis SUMO E3 ligase SIZ1 controls phosphate deficiency responses. Proc. Natl. Acad. Sci. USA 102:7760-7765. [PMC free article] [PubMed]
32. Moseley, J. L., T. Allinger, S. Herzog, P. Hoerth, E. Wehinger, S. Merchant, and M. Hippler. 2002. Adaptation to Fe-deficiency requires remodeling of the photosynthetic apparatus. EMBO J. 21:6709-6720. [PMC free article] [PubMed]
33. Nicholas, K. B., H. B. Nicholas, Jr., and D. W. Deerfield II. 1997. GeneDoc: analysis and visualization of genetic variation. EMBNEW.NEWS 4:14.
34. Niyogi, K. K. 1999. Photoprotection revisited: genetic and molecular approaches. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50:333-359. [PubMed]
35. Ogawa, N., J. DeRisi, and P. O. Brown. 2000. New components of a system for phosphate accumulation and polyphosphate metabolism in Saccharomyces cerevisiae revealed by genomic expression analysis. Mol. Biol. Cell 11:4309-4321. [PMC free article] [PubMed]
36. Oshima, Y. 1997. The phosphatase system in Saccharomyces cerevisiae. Genes Genet. Syst. 72:323-334. [PubMed]
37. Parsons, H. L., J. Y. Yip, and G. C. Vanlerberghe. 1999. Increased respiratory restriction during phosphate-limited growth in transgenic tobacco cells lacking alternative oxidase. Plant Physiol. 121:1309-1320. [PMC free article] [PubMed]
38. Peltier, G., and L. Cournac. 2002. Chlororespiration. Annu. Rev. Plant Biol. 53:523-550. [PubMed]
39. Quisel, J. D., D. D. Wykoff, and A. R. Grossman. 1996. Biochemical characterization of the extracellular phosphatases produced by phosphorus-deprived Chlamydomonas reinhardtii. Plant Physiol. 111:839-848. [PMC free article] [PubMed]
40. Raghothama, K. G. 1999. Phosphate acquisition. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50:665-693. [PubMed]
41. Richard, C., H. Ouellet, and M. Guertin. 2000. Characterization of the LI818 polypeptide from the green unicellular alga Chlamydomonas reinhardtii. Plant Mol. Biol. 42:303-316. [PubMed]
42. Rubio, V., F. Linhares, R. Solano, A. C. Martin, J. Iglesias, A. Leyva, and J. Paz-Ares. 2001. A conserved MYB transcription factor involved in phosphate starvation signaling both in vascular plants and in unicellular algae. Genes Dev. 15:2122-2133. [PMC free article] [PubMed]
43. Ruiz, F. A., N. Marchesini, M. Seufferheld, Govindjee, and R. Docampo. 2001. The polyphosphate bodies of Chlamydomonas reinhardtii possess a proton-pumping pyrophosphatase and are similar to acidocalcisomes. J. Biol. Chem. 276:46196-46203. [PubMed]
44. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
45. Schachtman, D. P., R. J. Reid, and S. M. Ayling. 1998. Phosphorus uptake by plants: from soil to cell. Plant Physiol. 116:447-453. [PMC free article] [PubMed]
46. Shimogawara, K., D. D. Wykoff, H. Usuda, and A. R. Grossman. 1999. Chlamydomonas reinhardtii mutants abnormal in their responses to phosphorus deprivation. Plant Physiol. 120:685-694. [PMC free article] [PubMed]
47. Shin, H., H. S. Shin, G. R. Dewbre, and M. J. Harrison. 2004. Phosphate transport in Arabidopsis: Pht1;1 and Pht1;4 play a major role in phosphate acquisition from both low- and high-phosphate environments. Plant J. 39:629-642. [PubMed]
48. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673-4680. [PMC free article] [PubMed]
49. Tottey, S., M. A. Block, M. Allen, T. Westergren, C. Albrieux, H. V. Scheller, S. Merchant, and P. E. Jensen. 2003. Arabidopsis CHL27, located in both envelope and thylakoid membranes, is required for the synthesis of protochlorophyllide. Proc. Natl. Acad. Sci. USA 100:16119-16124. [PMC free article] [PubMed]
50. Vogel, K., and A. Hinnen. 1990. The yeast phosphatase system. Mol. Microbiol. 4:2013-2018. [PubMed]
51. Weger, H. G., and G. S. Espie. 2000. Ferric reduction by iron-limited Chlamydomonas cells interacts with both photosynthesis and respiration. Planta 210:775-781. [PubMed]
52. Wu, P., L. Ma, X. Hou, M. Wang, Y. Wu, F. Liu, and X. W. Deng. 2003. Phosphate starvation triggers distinct alterations of genome expression in Arabidopsis roots and leaves. Plant Physiol. 132:1260-1271. [PMC free article] [PubMed]
53. Wykoff, D. D., J. P. Davies, A. Melis, and A. R. Grossman. 1998. The regulation of photosynthetic electron transport during nutrient deprivation in Chlamydomonas reinhardtii. Plant Physiol. 117:129-139. [PMC free article] [PubMed]
54. Wykoff, D. D., A. R. Grossman, D. P. Weeks, H. Usuda, and K. Shimogawara. 1999. Psr1, a nuclear localized protein that regulates phosphorus metabolism in Chlamydomonas. Proc. Natl. Acad. Sci. USA 96:15336-15341. [PMC free article] [PubMed]
55. Zer, H., and I. Ohad. 2003. Light, redox state, thylakoid-protein phosphorylation and signaling gene expression. Trends Biochem. Sci. 28:467-470. [PubMed]
56. Zhang, Z., J. Shrager, M. Jain, C. W. Chang, O. Vallon, and A. R. Grossman. 2004. Insights into the survival of Chlamydomonas reinhardtii during sulfur starvation based on microarray analysis of gene expression. Eukaryot. Cell 3:1331-1348. [PMC free article] [PubMed]

Articles from Eukaryotic Cell are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...